Animal treatment: Male Sprague-Dawley rats (300 $\pm$ 20 g body weight) were obtained from the experimental animal centre of Zhejiang Chinese Medical University [SCXK (Yu)-2005-3001], Zhejiang Province, P.R. China. The animal study was reviewed and approved by Animal Ethical and Welfare Committee of ZCMU. Animals were housed under standard conditions of temperature, humidity, and light and provided with plentiful food and water. All procedures involving animals were conducted in accordance with the regulations of experimental animal administration issued by the State Committee of Science and Technology of People’s Republic of China. After a week of adaptive feeding, rats were randomly divided into control groups and three LCD-treated groups. Rats were administered LCD decoction by gavage for 3 days at high (72.6 g/kg.bw), middle (36.3 g/kg.bw) and low (18.1 g/kg.bw) doses, respectively, which is listed as grams per kilogram rat body weight, to prepare high (H), middle (M) and low (L) doses of LCD serum. The doses were based on the conversion from human to rat according to clinical application. The untreated rats were administered distilled water accordingly. The dosages were calculated based on the weight of rats, which were measured daily.

LCD serum sample preparation: Blood was collected from the orbital vein of each rat on the 3rd day 1 h after the last drug administration, allowed to clot for approximately 45 min at room temperature and then centrifuged at 3000 rpm for 10 min. The obtained LCD serum samples were prepared for analyses as follows: 400 $\mu$L of cold methanol were added to 100 $\mu$L of serum for protein precipitation, which was performed in an ice-water bath. After vortex mixing for 30 s and centrifugation at 12,000 rpm at 4 °C for 10 min, the supernatant was collected and filtered through a 0.45 $\mu$m filter and stored at –80 °C for further analysis.

We integrated HPLC-MS analysis with molecular docking to further validate the interactions between CASP3 target and LCD. Different doses of LCD treated serum were prepared and identified with HPLC/MS analysis. LCD treated-rat serum were identified using an ion trap mass spectrometer equipped with an electrospray ionization (ESI) source (Thermo Fisher, USA) under the control of Xcalibur software (version 1.4, Thermo Fisher, USA) [16]. Chromatographic separation was performed using reverse-phase HPLC on a Dionex C18 column (4.6 $\times$ 250 mm id, particle size 5.0 $\mu$m). The mobile phase consisted of 0.1% formic acid (A) and acetonitrile (B) using the following 25 min gradient elution program: 0 min to 5 min, 80% A; 5 min to 10 min, 67% A; 10 min to 15 min, 66% A; and 15 min to 25 min, 60% A. The injection volume was 2 $\mu$L at a flow rate of 0.2 mL/min. The detection wavelength was set to 325 nm at a column temperature of 25 °C.

The HPLC eluents were analyzed with a mass spectrometer (Bruker, MA, USA) connected to an Ultimate 3000 HPLC system (Thermo Fisher Scientific, Wilmington, DE, USA) via an ESI interface. High purity nitrogen was used as the nebulizer and auxiliary gas, and argon was the collision gas. The Q-TOF-MS was operated in positive ion mode with a spray voltage of 4.5 KV and a collision-induced dissociation voltage of 25 V.

To further predict the possible component-target interaction, we used molecular docking simulation in this study. Structure information of the components was obtained from the NBCI PubChem database (https://pubchem.ncbi.nlm.nih.gov/). The three dimensional (3D) structures were optimized by Chem3D (ChemDraw Professional software 17.0, Cambridge, USA) and saved in MOL2 format. We obtained the 3D structure of protein target from the RCSB Protein Data Bank (PDB, https://www.rcsb.org/) [19] in PDB format and then used PyMOL software (Schrodinger, Inc., New York, USA) to remove water and separate ligand from the 3D protein structure. Finally, PDB format was saved as PDBQT files, further select Grid Box’s range in AutoDock software and run Autodock vina (http://vina.scripps.edu/) for virtual molecular docking [20].

Jurkat T cell viability was evaluated using a CCK-8 assay in triplicate [21, 22]. Briefly, cells (2 $\times$ 10${}^4$ cells/well) were cultured in 96-well plates with RPMI 1640 medium (10% foetal bovine serum) in a humidified atmosphere containing 5% CO${}_2$ at 37 °C. After adhesion, control and treated cells were treated with or without different doses of LCD serum (H, M and L) along activated with $\beta$-oestradiol (300 pg/mL) for 24 h and then incubated for 2 h with 10 $\mu$L of CCK-8 at 37 °C in the dark. After discarding cell supernatants, we detected the absorbance at 450 nm using a microplate spectrophotometer reader (Bio-Teck Instruments, Winooski, VT, USA).

Jurkat cells (2 $\times$ 10${}^6$ cells) cultured in 6-well plates were incubated with different doses of LCD serum (H, M and L) along with $\beta$-oestradiol (300 pg/mL) for 24 h. Following trypsinization, cells were washed twice with cold PBS, resuspended in binding buffer (1 $\times$ 10${}^6$ cells/mL) and 100.0 $\mu$L of each sample were transferred to separate 5 ml tubes. Cells were stained with 5 $\mu$L of Annexin V-FITC and 15 $\mu$L of propidium iodide (20 $\mu$g/mL) and then incubated in the dark at room temperature for 15 min. The stained cells were resuspended in 400 $\mu$L of 1x binding buffer. The cells were analysed using a FACS Vantage flow cytometer with Cell Quest software (Becton Dickinson, NJ, USA). The excitation wavelength was 488 nm and the emission wavelength was 670 nm.

Expression levels of the CASP3 mRNA in T cells were analysed using quantitative real-time PCR (RT–PCR). Briefly, Jurkat cells (1 $\times$ 10${}^6$ cells/well) cultured in 12-well plates were incubated in the absence or presence of different doses of LCD serum (H, M, and L) along with $\beta$-oestradiol (300 pg/mL) for 24 h. Then, total RNA was extracted from the cells using Trizol reagent according to the manufacturer’s instructions (Invitrogen, CA, USA), and the concentration of RNA was determined using a NanoDrop 2000 spectrophotometer (Thermo Fisher Scientific, Wilmington, DE, USA). The RNA was converted to cDNAs using a cDNA Reverse Transcription Kit (Bio-Rad, CA, USA). PCR amplification of the cDNA aliquots was performed using SYBR Green PCR master mix (Bio-Rad, CA, USA). Primers for CASP3 and $\beta$-actin were synthesized by Sangon Biotech (Shanghai, China). Forward and reverse primers for CASP3 and $\beta$-actin were 5´- TGCAGACGCGGCTCCTAGCG-3´ and 5´-AACCAGAGCGCCGAGTGTGAG-3´, 5´- CGAGCACAGAGCCTCGCCTT -3´ and 5´- ACATGCCGGAGCCGTTGTCG -3´, respectively (5´-3´ direction). Reaction was performed in PCR buffer with a total final volume of 10 $\mu$L under the following conditions: 15 s of denaturation at 94 °C, 10 s of annealing at 60 °C, and 10 s elongation at 72 °C. PCR cycles were repeated 45 times. All real-time PCR experiments were run in quadruplicate. The mRNA expression levels of CASP3 were analysed using the 2${}^{-\mathrm{\Delta }\mathrm{\Delta }\text{Ct}}$ method following normalization to $\beta$-actin.

Expression levels of the CASP3 protein were measured using western blot analysis. Briefly, Jurkat cells (1 $\times$ 10${}^6$ cells/well) cultured in 12-well plates were treated with different doses of LCD serum (H, M and L) along with $\beta$-oestradiol (300 pg/mL) at 37 °C in a 5% CO${}_2$ atmosphere. After 24 h, the cells were harvested, washed with PBS and lysed for 30 min on ice. Proteins in the lysates were quantified using a BCA protein assay kit. Subsequently, proteins (60 $\mu$g) were electrophoretically separated on 10% SDS-PAGE gels and then transferred to NC membranes. Membranes were blocked with 5% skim milk in PBS-T for 1 h at room temperature and then incubated with primary antibodies against CASP3 and $\beta$-actin at 4 °C overnight. After washing the protein blots with PBS-T, they were incubated with secondary antibodies at room temperature for 1 h. The positive bands in the blots were detected with an enhanced chemiluminescence reagent (Thermo Scientific, Waltham, MA, USA). The relative intensities of the protein bands were measured, and $\beta$-actin was used as an internal standard control.

The results are presented as the means $\pm$ standard deviations and were analysed using GraphPad Prism 5. One-way ANOVA was used to determine the statistical significance of differences between the values of various experimental groups.

To further predict CASP3 target of LCD treating SLE, we integrated HPLC-MS analysis with molecular docking. Different doses of LCD treated serum were prepared and identified with HPLC/MS analysis. According to the conditions described above, two components of ferulic acid (Fig. 2a) and isoferulic acid (Fig. 2b) in LCD serum were tentatively identified and showed better sensitivity in the positive ESI-MS analysis (Supplementary Table 4), which yielded protonated molecules at m/z 195. Based on comparisons of blank serum spiked reference standards (Fig. 2c) and LCD treated serum samples (Fig. 2d), we further performed a quantitative analysis including accuracy and precision. The linear-regression line of standards was shown in Table 2. Under the analytical conditions, the results of accuracy, intra-day precision and inter-day precision were shown in Supplementary Table 5. The specificity of the method was further supported based on the comparisons of chromatograms, retention times and reference standards. Based on the linear regression curves of standards, the contents of ferulic acid in high, middle and low doses of LCD serum (H, M and L) were 10.18 $\pm$ 0.36 $\mu$g/mL, 6.41 $\pm$ 0.25 $\mu$g/mL, and 5.88 $\pm$ 0.16 $\mu$g/mL and the contents of isoferulic acid were 42.23 $\pm$ 0.96 $\mu$g/mL, 34.28 $\pm$ 0.66 $\mu$g/mL, and 27.56 $\pm$ 0.46 $\mu$g/mL, respectively. A high dose of LCD serum (H) contained significantly higher amounts of ferulic acid and isoferulic acid than a middle dose (M) or a low dose (L).

Fig. 2.

Positive MS${}^{\mathrm{2}}$ spectra and Chromatographic profil for two compounds of ferulic acid and isoferulic acid. (a) Chemical structures and positive MS${}^2$ spectra for ferulic acid and b. isoferulic acid. (b) Chemical structures and positive MS${}^2$ spectra for isoferulic acid. (c) Blank serum spiked with reference standards of ferulic acid and isoferulic acid. (d) LCD-treated rat serum samples: (1) ferulic acid (2) isoferulic acid.

Molecular docking combined with molecular dynamics simulation were employed to reveal the detailed interactions between main compounds and potential protein targets. The potential target CASP3 with high degree and its main compounds of ferulic acid and isoferulic acid were selected for molecular docking based on the above results. Binding energies between CASP3 and compounds of ferulic acid and isoferulic acid were calculated in AUTODOCK vina. The binding energy less than zero mean spontaneous binding of targets and compounds. Lower docking energy indicating a stronger binding affinity and higher docking score. The lower the energy is, the binding is stable. It is generally accepted that the binding energy less than –5.0 kcal/mol was selected as the screening criteria. As shown in Fig. 3, ferulic acid (Fig. 3a) and isoferulic acid (Fig. 3b) buried in the binding pocket of the CASP3 target protein formed by amino-acid residues based on the hydrogen and hydrophobic interactions. Both compounds of ferulic acid and isoferulic acid bind to CASP3, but their potency were different, despite their similar structures. Three key amino acids, ARG286, GLU240 and GLU324 were found in the CASP3 (PDBID:1RHJ) - ferulic acid complex. Four key amino acids, ARG179, ARG341, GLN283 and SER236 were found in the CASP3 (PDBID:1RHU) - isoferulic acid complex. The binding energies of ferulic acid and isoferulic acid compounds with CASP3 were –5.8 kcal/mol and –5.5 kcal/mol, respectively, which showed good docking affinity. The results suggested that ferulic acid and isoferulic acid exhibited certain activity and CASP3 could be considered as an effective target of LCD in the treatment of SLE. LCD might exert protective effects against SLE through inducing CASP3 activities.

Fig. 3.

The molecular docking of CASP3 target with components of ferulic acid and isoferulic acid. (a) Ferulic acid–CASP3. (b) Isoferulic acid-CASP3.

We activated Jurkat T cells with $\beta$-oestradiol (300 pg/mL) and treated with different doses of LCD serum (H, M and L) for 24 h to study the potential anti-proliferation effect. The cell proliferation inhibitory rate was measured with a CCK-8 assay. As shown in Fig. 4a, the absorbance of the control group at OD450 was the highest of the four groups, indicating that $\beta$-oestradiol (300 pg/mL) induced the activation of Jurkat T cells. The viability of T cells was reduced after treatment with different doses of LCD serum (H, M and L) compared with the control group (p 0.05 or p 0.01). The inhibition rates were 54.6 %, 42.3 % and 14.3 %, respectively, after treatment with the three doses of LCD serum (H, M and L) as shown in Fig. 4b. The highest inhibition rate was observed after treatment with a high dose of LCD serum, 54.6 % (p 0.01). Based on these results, LCD exerted an anti-proliferative effect on Jurkat T cells, which may be attributed to their apoptosis induction.

Fig. 4.

Effects of LCD-treated rat serum on the inhibition rate of Jurkat T cells determined using the CCK-8 assay. Absorbance at OD450 (a) and T cell inhibition rate after treatment with different doses of LCD serum (H, M and L) (b). Values are presented as the mean $\pm$ SD. *p 0.05 and **p 0.01 compared with untreated control and LCD-treated cells. One-way analysis of variance (ANOVA) was used followed by Dunnett’s tests (n = 3).

To explore whether LCD serum induced apoptosis in Jurkat T cells, we stained and detected apoptotic cells with annexin V-FITC and PI. The percentage of apoptotic cells was quantified using flow cytometry and an annexin V/PI Apoptosis Assay Kit. Very few apoptotic cells were observed in the untreated control group (Fig. 5a). In contrast, the percentage of apoptotic cells increased gradually by 6.8%, 16.5% and 26.8% (Fig. 5b–d) after treatment with different doses of LCD serum (L, M and H) along with $\beta$-oestradiol (300 pg/mL), which were all significantly different from those of the untreated cells.

Fig. 5.

Representative flow cytometry profile of Jurkat T cells treated without (a) or with low (b), middle (c) and high (d) doses of LCD-treated rat serum along with $\beta$-oestradiol (300 pg/mL) for 24 h in vitro. Cells were stained with annexin V/PI and then analysed on a flow cytometer. A dual-parameter dot plot of annexin V-FITC fluorescence (x axis) versus PI fluorescence (y axis) is shown in logarithmic fluorescence intensity. Quadrants: lower left indicates live cells (-FITC), lower right indicates early apoptotic cells (+FITC), and upper right indicates late apoptotic cells (+PI and +FITC).

The early and late apoptosis rates increased gradually after treatment with low, middle and high doses of LCD serum (Supplementary Fig. 1). The increased apoptotic rates may be associated with LCD-induced apoptosis. The annexin V/PI double staining assay confirmed that LCD induces apoptosis in Jurkat T cells by initiating the apoptotic cascade.

We detected the expression of one important apoptosis-related protein, CASP3, using western blot analysis to study the potential mechanism by which LCD induced apoptosis. CASP3 is one of the executioner caspases activated by proteolytic cleavage during apoptosis. The levels of $\beta$-actin were monitored to ensure equal loading of protein samples. As shown in Fig. 6, compared with the control group treated with $\beta$-oestradiol (300 pg/mL), treatment with different doses of LCD serum (L, M and H) along with $\beta$-oestradiol increased the expression levels of the CASP3 mRNA (p 0.05, p 0.001 and p 0.001). Similarly, LCD serum (L, M and H) upregulated CASP3 protein levels (p 0.05, p 0.001 and p 0.001). The increased level of cleaved CASP3 showed that LCD induced apoptosis in Jurkat T cells.

Fig. 6.

Effects of LCD-treated rat serum on the caspase-3 (CASP3) mRNA and protein expression levels in Jurkat T cells. Cells were seeded in 6-well plates and treated with different doses of LCD serum (H, M and L) along with $\beta$-oestradiol (300 pg/mL) for 24 h. The protein levels were analysed using western blot analysis. $\beta$-actin was used as internal control. (a) Western blots showing levels of the $\beta$-actin and CASP3 proteins. (b) Expression level of the CASP3 mRNA. (c) Expression levels of the CASP3 protein. Values are presented as the means$\pm$ SD. **p 0.05, **p 0.01 and ***p 0.001 compared with untreated control and LCD-treated cells. One-way analysis of variance (ANOVA) was used followed by Dunnett’s test (n = 3).